CN106323167B - A kind of intelligent scanning on-line measurement system and measurement method based on image recognition - Google Patents
A kind of intelligent scanning on-line measurement system and measurement method based on image recognition Download PDFInfo
- Publication number
- CN106323167B CN106323167B CN201610704368.5A CN201610704368A CN106323167B CN 106323167 B CN106323167 B CN 106323167B CN 201610704368 A CN201610704368 A CN 201610704368A CN 106323167 B CN106323167 B CN 106323167B
- Authority
- CN
- China
- Prior art keywords
- measurement
- scanning
- robot
- ccd camera
- ecu
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
- G01B11/005—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates coordinate measuring machines
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of intelligent scanning on-line measurement system based on image recognition, it is related to industrial automation, including image recognition locating module, six-DOF robot laser scanning measurement module, measurement control module, wherein image recognition locating module includes CCD camera trigger switch, CCD camera, CCD camera bracket, CCD camera electronic control unit ECU；Measuring control module includes central processing unit, measuring system controller, robot measurement pose ECU, measuring head ECU；Six-DOF robot laser scanning measurement module includes six degree of freedom measurement robotic arm, robot measurement Position and attitude sensor, robot measurement joint motor, laser scanning measuring head, measurement trigger switch.The invention also discloses the intelligent scanning On-line Measuring Methods based on image recognition, including establish intelligent scanning policy database；Image recognition simultaneously carries out initial alignment；Adaptively scanning survey；Accurate secondary positioning during scanning survey；Scanning strategy during scanning survey updates.
Description
Technical field
The present invention relates to scan-type On-line Measuring Methods, refer in particular to a kind of intelligent scanning on-line measurement based on image recognition
System and measurement method.
Background technique
With the emergence of novel sensor, non-cpntact measurement is theoretical to be constantly improve, so that using non-cpntact measurement
Technology realizes that the on-line measurement of free form surface is possibly realized.It is non-contact to swash compared with traditional template measuring technique, CMM technology
Optical scanning detection technique has many advantages, such as that contactless, speed is fast for detection, data volume is big, in reverse-engineering, quality testing, virtual existing
The fields tool such as real has great advantage, and is widely deployed in production and life.
With industry 4.0 and internet+etc. concepts proposition, mass customization (OEM) production overturning thousand one
The homogeneous product of rule, factory also become fewer and fewer peopleization, unmanned, automation and intelligence.And the gentle life of Automated water
The raising for producing efficiency directly translates into the raising of production line beat, how to realize that the quality testing of product will be modern under fast beat
The critical issue of a period of time afterwards.
The quality management that current automotive industry and components manufacturer are directed to product mainly uses cubing and CMM sampling Detection
Method, or use on-line checking robotic gripper laser feeler, get measurement ready online for key point progress, and be aided with
The mode of CMM timing sampling observation carries out quality management.It is required to using the method for cubing and CMM sampling Detection for each part to be measured
Dedicated cubing is made, the cost of quality management is not only increased, and detection efficiency is low, cannot achieve on-line checking；Online
Robot key point on-line checking mode is detected, the detection of certain key points is can be only done, is unable to satisfy the comprehensive quality of product
Management requires, and especially for components production, can not comprehensively reflect part situation for small part critical point detection, together
When due to introducing robot measurement and dedicated CMM measuring system greatly improve production cost, thus significantly limit
Application of the robot on-line checking in components production.
Therefore, those skilled in the art is dedicated to developing a kind of intelligent scanning on-line measurement system based on image recognition
And measurement method, realize for multi items, fast beat, high-volume personalized customization components automated production online inspection
It surveys.
Summary of the invention
In view of the above drawbacks of the prior art, the technical problem to be solved by the present invention is to how for multi items, fast
Beat, high-volume personalized customization components automated production on-line checking, realize be based on part feature and measurement request
Adaptive laser scanning measurement, simplify measuring device, improve measurement efficiency and comprehensive.
To achieve the above object, the present invention provides a kind of intelligent scanning on-line measurement system based on image recognition, packet
Image recognition locating module, six-DOF robot laser scanning measurement module, measurement control module are included, wherein described image is known
Other locating module includes CCD camera trigger switch, CCD camera, CCD camera bracket, CCD camera electronic control unit ECU；It is described
Measuring control module includes central processing unit, measuring system controller, robot measurement pose ECU, measuring head ECU；Described six
Degree of freedom robot laser scanning measurement module includes six degree of freedom measurement robotic arm, robot measurement Position and attitude sensor, measurement
Joint of robot motor, laser scanning measuring head, measurement trigger switch.
Further, the CCD camera trigger switch is located on production line, the CCD camera beside production line
The CCD camera is supported in above production line by bracket.
Further, the measuring system controller acquires the robot measurement pose ECU and measuring head ECU
Pose data and measurement data be sent to the central processing unit, while the control command of the central processing unit being sent to
The robot measurement pose ECU and measuring head ECU.
Further, the measurement trigger switch is located at production line measurement station, and the laser scanning head is located at described
Six degree of freedom measures the end of robotic arm, and the robot measurement joint motor and the robot measurement Position and attitude sensor are located at
Each joint of six degree of freedom measurement robotic arm.
The intelligent scanning On-line Measuring Method based on image recognition that the present invention also provides a kind of, comprising the following steps:
Step 1 establishes intelligent scanning policy database；
Step 2, image recognition simultaneously carry out initial alignment；
Step 3, adaptively scanning survey；
Accurate secondary positioning during step 4, scanning survey；
Scanning strategy during step 5, scanning survey updates.
Further, in step 1, further include
The three-dimensional part model of all kinds of online production on production line is imported central processing unit by step 1.1；
Step 1.2, central processing unit will be to according to the aspect of model and measurement request of the threedimensional model of each part to be measured
The threedimensional model for surveying part is converted to the target measurement point cloud of gridding domain type；
Step 1.3 makes the adaptive scanning strategy for meeting the aspect of model and measurement request according to target measurement point cloud.
Further, in step 2, further include
Step 2.1, when part weld complete rearward measurement station it is mobile during, when by CCD camera trigger switch,
CCD camera trigger switch, which is triggered, sends CCD camera trigger signal to CCD camera, and then controls CCD camera and clap part
It takes the photograph；
The three-dimensional part image of the part of acquisition is transmitted to CCD camera ECU, CCD camera ECU warp by step 2.2, CCD camera
Processing is crossed, three-dimensional part image is converted into the part 3-dimensional digital signal with coordinate position and is transmitted to central processing unit；
Step 2.3, central processing unit according to the feature of part 3-dimensional digital signal identify corresponding part and with importing
Corresponding three-dimensional part model is corresponding, and identifies the corresponding coordinate information of three-dimensional part model elements of fix, realizes that image is known
Other function；
Step 2.4, central processing unit sit the corresponding elements of fix of the part 3-dimensional digital signal obtained by image information
Information is marked, the initial coordinate at measurement position is reached in conjunction with robot measurement initial coordinate position and part, obtains measuring machine
Device people is directed to the measurement coordinate of part to be measured；
It is step 2.5, the corresponding elements of fix coordinate information of part 3-dimensional digital signal is corresponding with target measurement point cloud
Element coordinate information carries out optimal fitting, to realize image initial positioning function, Robot Scanning is measured coordinate and model
Target measurement point cloud coordinate unification, provides the reference data of scanning for subsequent adaptive scanning.
Further, in step 3, further include
Step 3.1, when load part retinue platform reach measurement station when, positioned at measurement station at measurement triggering opens
Pass will be triggered, and measurement trigger signal be issued to measuring system controller, after measuring system controller receives measurement trigger signal
Start process of measurement；
The adaptive scanning strategy and measurement machine that step 3.2, measuring system controller are issued according to central processing unit
The robot pose feedback signal that people's pose ECU is sent generates robot Pose Control signal and is transmitted to robot measurement pose ECU,
The pose signal that robot measurement pose ECU is surveyed according to robot Pose Control signal and robot measurement Position and attitude sensor
Generating joint of robot motor control signal reaches laser scanning head to control the pose of robot measurement joint motor
Scan position；
Step 3.3, when laser scanning head reach scan position when, measuring system controller will be received from robot measurement
The robot pose feedback signal that pose ECU is sent, measuring system controller sending scan control signal to measuring head ECU, by
Measuring head ECU generates scanning signal, so that controlling laser scanning head completes scanning；
Laser signal from laser scanning head is converted to position digital signal and is transmitted to measurement by step 3.4, measuring head ECU
System controller, measuring system controller generate measurement pointcloud number according to position digital signal and robot pose feedback signal
According to passing central processing unit back；
The target measurement point that step 3.5, central processing unit convert the threedimensional model of measurement pointcloud data and workpiece for measurement
Cloud together in real time shows measurement result.
Further, in step 4, further include
Step 4.1, starting measurement after measuring system controller receives the measurement trigger signal of measurement trigger switch sending
Program, measuring system controller first control measuring system and complete the corresponding elements of fix scanning of three-dimensional part model, central processing
After device receives the scanning survey point cloud data of initial alignment element, by the scanning survey point cloud data and target of initial alignment element
Element coordinate information is corresponded in measurement pointcloud carries out quadratic fit；
Step 4.2 precisely aligns Robot Scanning measurement coordinate and simulated target measurement pointcloud coordinate, is subsequent adaptive
It should scan and the exact references benchmark of scanning is provided.
Further, in steps of 5, further include
After step 5.1, central processing unit receive measurement pointcloud data, surveyed in conjunction with the threedimensional model and target of part to be measured
Amount point cloud, calculates new object to be measured measurement pointcloud, makes new adaptive scanning strategy and be transmitted to measuring system controller；
Step 5.2, measuring system controller measure to realize scanning strategy according to new adaptive scanning strategy
It is adjusted with the dynamic of scanning process, until completing the measurement of part to be measured；
Step 5.3, robot measurement return to initial position.
Intelligent scanning type on-line measurement system of the present invention based on image recognition, as shown in Figure 1, including that image is known
Other locating module, six-DOF robot laser scanning measurement module, measurement control module, further include welding and assembling production lines, are located at
On welding and assembling production lines retinue platform, positioned at retinue platform on pallet pallet, the fixture on pallet pallet and
Part on fixture, wherein image recognition locating module is by CCD camera trigger switch, CCD camera, CCD camera bracket, CCD phase
Electromechanical sub-control unit (ECU) composition, CCD camera trigger switch are located on production line, the CCD camera branch beside production line
CCD camera is supported in above production line by frame；Six-DOF robot laser scanning measurement module measures machine by six degree of freedom
Arm, robot measurement Position and attitude sensor, robot measurement joint motor, laser scanning measuring head, measurement trigger switch, measurement touching
Hair switch is located at production line measurement station, and laser scanning head is located at the end of six degree of freedom measurement robotic arm, robot measurement
Joint motor and robot measurement Position and attitude sensor are located at each joint of six degree of freedom measurement robotic arm；Measure control module by
Central processing unit, measuring system controller, robot measurement pose ECU, measuring head ECU composition.
Intelligent scanning type on-line measurement system intelligent scanning of the present invention based on image recognition detects control method:
1) intelligent scanning policy database, process are established are as follows: first by zero of all kinds of online production on production line
Part threedimensional model imports central processing unit, the aspect of model and measurement of the central processing unit according to the threedimensional model of each part to be measured
It is required that the threedimensional model of part to be measured to be converted to the target measurement point cloud of gridding domain type, and then according to target measurement point cloud
Make the adaptive scanning strategy for meeting the aspect of model and measurement request.
2) image recognition and initial alignment, process are carried out are as follows: complete the mobile process of rearward measurement station when part welds
In, when by CCD camera trigger switch, CCD camera trigger switch, which is triggered, sends CCD camera trigger signal to CCD camera, into
And control CCD camera and part is shot, the three-dimensional part image of the part of acquisition is transmitted to CCD camera by subsequent CCD camera
Three-dimensional part image is converted to the part 3-dimensional digital signal with coordinate position and passed by ECU, CCD camera ECU by processing
To central processing unit, central processing unit is according to part 3-dimensional digital signal, three-dimensional part model, first basis with coordinate position
The feature of part 3-dimensional digital signal identifies corresponding part and corresponding three-dimensional part model is corresponding to importing, and identifies this
The corresponding coordinate information of three-dimensional part model elements of fix realizes image identification function；Then zero will obtained by image information
The corresponding elements of fix coordinate information of part 3-dimensional digital signal is reached in conjunction with robot measurement initial coordinate position and part
Initial coordinate at measurement position obtains the measurement coordinate that robot measurement is directed to part to be measured；Finally again by three dimension of part
Element coordinate information is corresponded in the corresponding elements of fix coordinate information of word signal and target measurement point cloud carries out optimal fitting, thus
It realizes image initial positioning function, is subsequent by Robot Scanning measurement coordinate and simulated target measurement pointcloud coordinate unification
Adaptive scanning provides the reference data of scanning.
3) adaptively scanning survey, process are as follows: when the retinue platform for loading part reaches measurement station, be located at measurement
Measurement trigger switch at station will be triggered, and issue measurement trigger signal, measuring system controller to measuring system controller
Start process of measurement, the adaptive scanning plan that measuring system controller is issued according to central processing unit after receiving measurement trigger signal
The robot pose feedback signal that summary and robot measurement pose ECU are sent generates robot Pose Control signal and is transmitted to measurement
Robot pose ECU, robot measurement pose ECU are according to robot Pose Control signal and robot measurement Position and attitude sensor
The pose signal surveyed generates joint of robot motor control signal to be made to control the pose of robot measurement joint motor
Laser scanning head reaches scan position；When laser scanning head reaches scan position, measuring system controller will be received from survey
The robot pose feedback signal that amount robot pose ECU is sent, then measuring system controller issues scan control signal extremely
Measuring head ECU generates scanning signal by measuring head ECU, so that controlling laser scanning head completes scanning；Measuring head ECU will come from
The laser signal of laser scanning head is converted to position digital signal and is transmitted to measuring system controller, and measuring system controller is according to position
It sets digital signal and robot pose feedback signal generates measurement pointcloud data and passes central processing unit back, central processing unit will be surveyed
Amount point cloud data is presented together with the target measurement point cloud that the threedimensional model of workpiece for measurement is converted, thus in real time by measurement result
It shows, realizes the adaptive scanning to workpiece for measurement.
4) the accurate secondary position fixing process during scanning survey are as follows: when measurement starts, when measuring system controller receives
Start process of measurement after the measurement trigger signal that measurement trigger switch issues, measuring system controller first controls measuring system completion
The corresponding elements of fix scanning of three-dimensional part model, central processing unit receive the scanning survey point cloud data of initial alignment element
Afterwards, element coordinate information will be corresponded in the scanning survey point cloud data of initial alignment element and target measurement point cloud carries out secondary intend
It closes, to realize the secondary precise locating function of laser scanning, Robot Scanning measurement coordinate and simulated target measurement pointcloud is sat
Mark precisely aligns, to provide the exact references benchmark of scanning for subsequent adaptive scanning.
5) the scanning strategy renewal process during scanning survey are as follows: after central processing unit receives measurement pointcloud data, knot
The threedimensional model and target measurement point cloud for closing part to be measured, calculate new object to be measured measurement pointcloud, make it is new from
It adapting to scanning strategy and is transmitted to measuring system controller, measuring system controller is measured according to new adaptive scanning strategy,
To realize that scanning strategy is adjusted with the dynamic of scanning process, until completing the measurement of part to be measured, subsequent robot measurement is returned
Return initial position.
Beneficial effects of the present invention are as follows:
1, the present invention completes components using six degree of freedom robot measurement online and is welded the measurement of production, not only can be real
Now automatic quick scanning survey improves detection efficiency and does not need to make special inspection simultaneously because scanning survey is completed online
Have fixed measured workpiece measure under line, greatly extends the scope of application of scanning survey, the resource for reducing measurement disappears
Consumption.
2, the present invention uses identification and initial alignment function based on image procossing, by measuring basis and target point before measurement
Cloud carries out coordinate alignment, and the analysis that may be implemented in measurement process is compared, and can formulate and measure according to target point cloud direct-on-line
Strategy, and dynamic tracking measurement；Measuring system is set to be adapted to the weldering of multi items welding production line multi items using image recognition
The detection of pipe fitting measures.
3, the present invention is according to the aspect of model and measurement request of workpiece for measurement threedimensional model by the threedimensional model of workpiece for measurement
It is converted to the target measurement point cloud of gridding domain type, and passes through the secondary of initial alignment based on image procossing and laser scanning
Precise positioning, realization target measurement point cloud coordinate are registrated with measurement pointcloud coordinate, can be made and be met workpiece for measurement model spy
The optimal scanning strategy for measurement request of seeking peace, and realize measurement adjustment in real time in measurement process, it realizes to workpiece for measurement
Three-dimensional adaptive scanning survey makes measurement more be bonded measurement object and measurement request.
It is described further below with reference to technical effect of the attached drawing to design of the invention, specific structure and generation, with
It is fully understood from the purpose of the present invention, feature and effect.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of a preferred embodiment of the invention；
Fig. 2 is the method flow diagram of a preferred embodiment of the invention；
In figure: 1 is CCD camera trigger switch, and 2 be CCD camera, and 3 be CCD camera bracket, and 4 measure machine for six degree of freedom
Arm, 5 be robot measurement Position and attitude sensor, and 6 be robot measurement joint motor, and 7 be laser scanning measuring head, and 8 is flat for retinue
Platform, 9 be pallet pallet, and 10 be fixture, and 11 be part, and 12 be welding and assembling production lines, and 13 be CCD camera electronic control unit
(ECU), 14 be central processing unit, and 15 be measuring system controller, and 16 be robot measurement pose ECU, and 17 be measuring head ECU,
18 be CCD camera trigger signal, and 19 be part image, and 20 be part 3-dimensional digital signal, and 21 be three-dimensional part model, and 22 be survey
Trigger signal is measured, 23 be adaptive scanning strategy, and 24 be measurement pointcloud data, and 25 be robot Pose Control signal, and 26 be machine
Device people's pose feedback signal, 17 be joint of robot motor control signal, and 28 be joint of robot motor feedback signals, and 29 be position
Appearance signal, 30 be scan control signal, and 31 be position digital signal, and 32 be scanning signal, and 33 be laser signal, and 34 be measurement touching
Hair switch.
Specific embodiment
The invention is described in further details with reference to the accompanying drawing.
As shown in Figure 1, the intelligent scanning type on-line measurement system of the present invention based on image recognition, including image are known
Other locating module, six-DOF robot laser scanning measurement module, measurement control module, further include welding and assembling production lines 12, position
In on the retinue platform 8 on welding and assembling production lines 12, the pallet pallet 9 on retinue platform 8, pallet pallet 9
Part 11 on fixture 10 and fixture 10, wherein image recognition locating module by CCD camera trigger switch 1, CCD camera 2,
CCD camera bracket 3, CCD camera electronic control unit (ECU) 13 form, and CCD camera trigger switch 1 is located on production line, is located at
CCD camera 2 is supported in above production line by the CCD camera bracket 3 beside production line；Six-DOF robot laser scanning measurement
Module measures robotic arm 4, robot measurement Position and attitude sensor 5, robot measurement joint motor 6, laser scanning by six degree of freedom
Measuring head 7, measurement trigger switch 34, measurement trigger switch 34 are located at production line measurement station, and laser scanning head 7 is located at six certainly
By the end of degree measurement robotic arm 1, robot measurement joint motor 6 and robot measurement Position and attitude sensor 7 are located at six degree of freedom
Measure each joint of robotic arm 1；Control module is measured by central processing unit 14, measuring system controller 15, robot measurement
Pose ECU16, measuring head ECU17 composition.
As shown in Fig. 2, the intelligent scanning type on-line measurement system intelligent scanning inspection of the present invention based on image recognition
Survey control method:
1) intelligent scanning policy database, process are established are as follows: first by zero of all kinds of online production on production line
Part threedimensional model 21 imports central processing unit 14, and central processing unit 14 is special according to the model of the threedimensional model 21 of each part to be measured
The threedimensional model 21 of part to be measured is converted to the target measurement point cloud of gridding domain type by measurement request of seeking peace, and then according to mesh
Mark measurement pointcloud makes the adaptive scanning strategy 23 for meeting the aspect of model and measurement request；
2) image recognition and initial alignment, process are carried out are as follows: complete the mobile mistake of rearward measurement station when part 11 welds
Cheng Zhong, when by CCD camera trigger switch 1, CCD camera trigger switch 1, which is triggered, sends CCD camera triggering letter to CCD camera 2
Numbers 18, and then control CCD camera 2 and shoot to part 11, subsequent CCD camera 2 is by the three-dimensional part drawing of the part 11 of acquisition
As 19 are transmitted to CCD camera ECU13, CCD camera ECU13 is converted to band coordinate position by processing, by three-dimensional part image 19
Part 3-dimensional digital signal 20 and be transmitted to central processing unit 14, central processing unit 14 is according to three dimension of part with coordinate position
Word signal 20, three-dimensional part model 21, first according to the feature of part 3-dimensional digital signal 20 identify corresponding part and with lead
The corresponding three-dimensional part model 21 entered is corresponding, and identifies the corresponding coordinate information of 21 elements of fix of three-dimensional part model, real
Existing image identification function；Then the corresponding elements of fix coordinate of part 3-dimensional digital signal 20 obtained by image information is believed
Breath reaches the initial coordinate at measurement position in conjunction with robot measurement initial coordinate position and part 11, obtains measurement machine
People is directed to the measurement coordinate of part 11 to be measured；Finally again by the corresponding elements of fix coordinate information of part 3-dimensional digital signal 20 with
Element coordinate information is corresponded in target measurement point cloud and carries out optimal fitting, so that image initial positioning function is realized, by robot
Scanning survey coordinate and simulated target measurement pointcloud coordinate unification, to provide the initial ginseng of scanning for subsequent adaptive scanning
Examine benchmark.
3) adaptively scanning survey, process are as follows: when the retinue platform for loading part 11 reaches measurement station, be located at and survey
Measurement trigger switch 34 at amount station will be triggered, and issue measurement trigger signal 22, measurement system to measuring system controller 15
System controller 15 starts process of measurement after receiving measurement trigger signal 22, and measuring system controller 15 is sent out according to central processing unit 14
The robot pose feedback signal 26 that adaptive scanning strategy 23 and robot measurement pose ECU16 out is sent generates machine
People's Pose Control signal 25 is transmitted to robot measurement pose ECU16, and robot measurement pose ECU16 is according to robot Pose Control
The pose signal 29 that signal 25 and robot measurement Position and attitude sensor 5 are surveyed generates joint of robot motor control signal 27,
To control the pose of robot measurement joint motor 6, laser scanning head 7 is made to reach scan position；When laser scanning head 7 reaches
When scan position, measuring system controller 15 is fed back the robot pose sent from robot measurement pose ECU16 is received
Signal 26, then measuring system controller 15 issues scan control signal 30 to measuring head ECU17, is generated by measuring head ECU17
Scanning signal 32, so that controlling laser scanning head 7 implements scanning；Measuring head ECU17 is by the laser signal from laser scanning head 7
33, which are converted to position digital signal 31, is transmitted to measuring system controller 15, and measuring system controller 15 is according to position digital signal 31
And robot pose feedback signal 26 generates measurement pointcloud data 24 and passes central processing unit 14 back, central processing unit 14 will measure
Point cloud data 24 is presented together with the target measurement point cloud that the threedimensional model 20 of workpiece for measurement is converted, in real time tie measurement
Fruit shows, and realizes the adaptive scanning to workpiece for measurement.
4) the accurate secondary position fixing process during scanning survey are as follows: when measurement starts, when measuring system controller 15 is received
Start process of measurement after the measurement trigger signal 22 issued to measurement trigger switch 34, measuring system controller 15 first controls measurement
System completes the scanning of the corresponding elements of fix of three-dimensional part model 21, and the scanning that central processing unit 14 receives initial alignment element is surveyed
After measuring point cloud data 24, element coordinate will be corresponded in the scanning survey point cloud data 24 of initial alignment element and target measurement point cloud
Information carries out quadratic fit, to realize the secondary precise locating function of laser scanning, Robot Scanning is measured coordinate and model
Target measurement point cloud coordinate precisely aligns, to provide the exact references benchmark of scanning for subsequent adaptive scanning.
5) the scanning strategy renewal process during scanning survey are as follows: central processing unit 14 receives measurement pointcloud data 24
Afterwards, in conjunction with the threedimensional model 21 of part 11 to be measured and target measurement point cloud, new object to be measured measurement pointcloud is calculated, is formulated
New adaptive scanning strategy 23 is transmitted to measuring system controller 15 out, and measuring system controller 15 is according to new adaptive scanning
Strategy 23 measures, to realize that scanning strategy 23 is adjusted with the dynamic of scanning process, until completing the survey of part 11 to be measured
Amount, subsequent robot measurement return to initial position.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that the ordinary skill of this field is without wound
The property made labour, which according to the present invention can conceive, makes many modifications and variations.Therefore, all technician in the art
Pass through the available technology of logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Scheme, all should be within the scope of protection determined by the claims.
Claims (7)
1. a kind of intelligent scanning on-line measurement system based on image recognition, which is characterized in that including image recognition locating module,
Six-DOF robot laser scanning measurement module, measurement control module, wherein described image identification locating module includes CCD phase
Machine trigger switch, CCD camera, CCD camera bracket, CCD camera electronic control unit ECU；During the measurement control module includes
Central processor, measuring system controller, robot measurement pose ECU, measuring head ECU；The six-DOF robot laser is swept
Retouching measurement module includes six degree of freedom measurement robotic arm, robot measurement Position and attitude sensor, robot measurement joint motor, laser
Scanning survey head, measurement trigger switch；
The measurement trigger switch is located at production line measurement station, and the laser scanning head is located at the six degree of freedom measuring machine
The end of device arm, the robot measurement joint motor and the robot measurement Position and attitude sensor are located at six degree of freedom measuring machine
Each joint of device arm；
The on-line measurement system uses a kind of intelligent scanning On-line Measuring Method based on image recognition, comprising the following steps:
Step 1 establishes intelligent scanning policy database；
Step 2, image recognition simultaneously carry out initial alignment；
Step 3, adaptively scanning survey；
Accurate secondary positioning during step 4, scanning survey；
Scanning strategy during step 5, scanning survey updates；
In step 2, further includes:
Step 2.1, when part weld complete rearward measurement station it is mobile during, when by CCD camera trigger switch, CCD
Camera trigger switch, which is triggered, sends CCD camera trigger signal to CCD camera, and then controls CCD camera and shoot to part；
The three-dimensional part image of the part of acquisition is transmitted to CCD camera ECU by step 2.2, CCD camera, and CCD camera ECU is at
Three-dimensional part image is converted to the part 3-dimensional digital signal with coordinate position and is transmitted to central processing unit by reason；
Step 2.3, central processing unit identify corresponding part and corresponding to importing according to the feature of part 3-dimensional digital signal
Three-dimensional part model is corresponding, and identifies the corresponding coordinate information of three-dimensional part model elements of fix, realizes image recognition function
Energy；
Step 2.4, central processing unit believe the corresponding elements of fix coordinate of the part 3-dimensional digital signal obtained by image information
Breath reaches the initial coordinate at measurement position in conjunction with robot measurement initial coordinate position and part, obtains robot measurement
For the measurement coordinate of part to be measured；
Step 2.5 will correspond to element in the corresponding elements of fix coordinate information of part 3-dimensional digital signal and target measurement point cloud
Coordinate information carries out optimal fitting, to realize image initial positioning function, Robot Scanning is measured coordinate and simulated target
Measurement pointcloud coordinate unification provides the reference data of scanning for subsequent adaptive scanning.
2. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that the CCD
Camera trigger switch is located on production line, and the CCD camera is supported in life by the CCD camera bracket beside production line
Above producing line.
3. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that the measurement
System controller sends the robot measurement pose ECU and measuring head ECU pose data acquired and measurement data
To the central processing unit, at the same by the control command of the central processing unit be sent to the robot measurement pose ECU and
The measuring head ECU.
4. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that in the base
In the step 1 of the intelligent scanning On-line Measuring Method of image recognition, further include
The three-dimensional part model of all kinds of online production on production line is imported central processing unit by step 1.1；
Step 1.2, central processing unit are according to the aspect of model of the threedimensional model of each part to be measured and measurement request by be measured zero
The threedimensional model of part is converted to the target measurement point cloud of gridding domain type；
Step 1.3 makes the adaptive scanning strategy for meeting the aspect of model and measurement request according to target measurement point cloud.
5. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that in the base
In the step 3 of the intelligent scanning On-line Measuring Method of image recognition, further include
Step 3.1, when load part retinue platform reach measurement station when, positioned at measurement station at measurement trigger switch will
It is triggered, issues measurement trigger signal to measuring system controller, measuring system controller starts after receiving measurement trigger signal
Process of measurement；
The adaptive scanning strategy and robot measurement position that step 3.2, measuring system controller are issued according to central processing unit
The robot pose feedback signal that appearance ECU is sent generates robot Pose Control signal and is transmitted to robot measurement pose ECU, measures
Robot pose ECU is generated according to the pose signal that robot Pose Control signal and robot measurement Position and attitude sensor are surveyed
Joint of robot motor control signal makes laser scanning head reach scanning to control the pose of robot measurement joint motor
Position；
Step 3.3, when laser scanning head reach scan position when, measuring system controller will be received from robot measurement pose
The robot pose feedback signal that ECU is sent, measuring system controller issue scan control signal to measuring head ECU, by measuring
Head ECU generates scanning signal, so that controlling laser scanning head completes scanning；
Laser signal from laser scanning head is converted to position digital signal and is transmitted to measuring system by step 3.4, measuring head ECU
Controller, measuring system controller generate measurement pointcloud data according to position digital signal and robot pose feedback signal and pass
Return central processing unit；
The target measurement point cloud one that step 3.5, central processing unit convert the threedimensional model of measurement pointcloud data and workpiece for measurement
It rises and in real time shows measurement result.
6. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that in the base
In the step 4 of the intelligent scanning On-line Measuring Method of image recognition, further include
Step 4.1 starts process of measurement after measuring system controller receives the measurement trigger signal that measurement trigger switch issues,
Measuring system controller first controls measuring system and completes the corresponding elements of fix scanning of three-dimensional part model, and central processing unit receives
After the scanning survey point cloud data of initial alignment element, by the scanning survey point cloud data of initial alignment element and target measurement point
Element coordinate information is corresponded in cloud carries out quadratic fit；
Step 4.2 precisely aligns Robot Scanning measurement coordinate and simulated target measurement pointcloud coordinate, adaptively sweeps to be subsequent
It retouches and the exact references benchmark of scanning is provided.
7. the intelligent scanning on-line measurement system based on image recognition as described in claim 1, which is characterized in that in the base
In the step 5 of the intelligent scanning On-line Measuring Method of image recognition, further include
After step 5.1, central processing unit receive measurement pointcloud data, in conjunction with threedimensional model and the target measurement point of part to be measured
Cloud calculates new object to be measured measurement pointcloud, makes new adaptive scanning strategy and is transmitted to measuring system controller；
Step 5.2, measuring system controller measure to realize scanning strategy with sweeping according to new adaptive scanning strategy
The dynamic adjustment for retouching process, until completing the measurement of part to be measured；
Step 5.3, robot measurement return to initial position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610704368.5A CN106323167B (en) | 2016-08-22 | 2016-08-22 | A kind of intelligent scanning on-line measurement system and measurement method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610704368.5A CN106323167B (en) | 2016-08-22 | 2016-08-22 | A kind of intelligent scanning on-line measurement system and measurement method based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106323167A CN106323167A (en) | 2017-01-11 |
| CN106323167B true CN106323167B (en) | 2019-06-07 |
Family
ID=57741858
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610704368.5A Active CN106323167B (en) | 2016-08-22 | 2016-08-22 | A kind of intelligent scanning on-line measurement system and measurement method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106323167B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI821027B (en) * | 2022-11-17 | 2023-11-01 | 英業達股份有限公司 | Component matching and reporting method |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106767517B (en) * | 2017-01-18 | 2019-03-01 | 重庆市计量质量检测研究院 | Trailing edge slot pattern automatic acquiring method and acquisition device in a kind of free form surface |
| CN106705847B (en) * | 2017-01-18 | 2019-04-05 | 重庆市计量质量检测研究院 | Gap geometric dimension measurement method and device in a kind of free form surface |
| CN107144236A (en) * | 2017-05-25 | 2017-09-08 | 西安交通大学苏州研究院 | A kind of robot automatic scanner and scan method |
| CN107598775A (en) * | 2017-07-27 | 2018-01-19 | 芜湖微云机器人有限公司 | It is a kind of by laser detect the method that simultaneously multi-axis robot is polished |
| CN107270833A (en) * | 2017-08-09 | 2017-10-20 | 武汉智诺维科技有限公司 | A kind of complex curved surface parts three-dimension measuring system and method |
| CN107576503A (en) * | 2017-09-01 | 2018-01-12 | 西北工业大学 | Aero-engine high accuracy assembly quality detector device people's system |
| CN108362223B (en) * | 2017-11-24 | 2020-10-27 | 广东康云多维视觉智能科技有限公司 | Portable 3D scanner, scanning system and scanning method |
| CN109839075A (en) * | 2017-11-29 | 2019-06-04 | 沈阳新松机器人自动化股份有限公司 | A kind of robot automatic measurement system and measurement method |
| CN108274092B (en) * | 2017-12-12 | 2020-08-21 | 北京石油化工学院 | Automatic groove cutting system and method based on three-dimensional vision and model matching |
| CN108274302B (en) * | 2017-12-25 | 2020-11-10 | 重庆近江智信汽车零部件有限公司 | Intelligent pipe making machine |
| CN108489361A (en) * | 2018-03-02 | 2018-09-04 | 昆山艾尔发计量科技有限公司 | A kind of logical only detecting system of workpiece with hole |
| CN110006361B (en) * | 2019-03-12 | 2021-08-03 | 精诚工科汽车系统有限公司 | Automatic part detection method and system based on industrial robot |
| CN110207619B (en) * | 2019-04-10 | 2022-01-07 | 北京航星机器制造有限公司 | Measuring system and method for carrying cooperative mechanical arm based on omnibearing mobile platform |
| CN110174346A (en) * | 2019-05-06 | 2019-08-27 | 江苏耀群工业技术有限公司 | Autoregistration strainer scan method and device |
| CN110244653B (en) * | 2019-06-13 | 2020-10-09 | 深圳市腾浩科技有限公司 | Method for planning machining trajectory of workpiece |
| CN111397507B (en) * | 2019-12-30 | 2021-10-01 | 宁波大正工业机器人技术有限公司 | Disc part laser calibration method |
| CN113379905A (en) * | 2021-07-16 | 2021-09-10 | 秦皇岛信能能源设备有限公司 | Wheel hub is 3D vision intelligent identification and control system for production line |
| CN113959352B (en) * | 2021-10-19 | 2022-09-09 | 上海交通大学 | Flexible point location direction identification system and method |
| CN115717868B (en) * | 2022-10-08 | 2023-11-07 | 北京航天计量测试技术研究所 | Real-time online three-dimensional automatic scanning measurement system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101044505A (en) * | 2004-07-23 | 2007-09-26 | 3形状股份有限公司 | Adaptiv 3d scanning |
| CN101236660A (en) * | 2008-03-06 | 2008-08-06 | 张利群 | Three-dimensional scanners and its three-dimensional model reconfiguration method |
| CN102663810A (en) * | 2012-03-09 | 2012-09-12 | 北京航空航天大学 | Full-automatic modeling approach of three dimensional faces based on phase deviation scanning |
| CN202523112U (en) * | 2012-04-09 | 2012-11-07 | 北京拓博尔机器人科技有限公司 | System for reconstructing three-dimensional model by robot carrying laser scanner |
| CN104036546A (en) * | 2014-06-30 | 2014-09-10 | 清华大学 | Method for carrying out face three-dimensional reconstruction at any viewing angle on basis of self-adaptive deformable model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104344795A (en) * | 2013-07-24 | 2015-02-11 | 鸿富锦精密工业（深圳）有限公司 | Point cloud profile measuring system and method |
-
2016
- 2016-08-22 CN CN201610704368.5A patent/CN106323167B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101044505A (en) * | 2004-07-23 | 2007-09-26 | 3形状股份有限公司 | Adaptiv 3d scanning |
| CN101236660A (en) * | 2008-03-06 | 2008-08-06 | 张利群 | Three-dimensional scanners and its three-dimensional model reconfiguration method |
| CN102663810A (en) * | 2012-03-09 | 2012-09-12 | 北京航空航天大学 | Full-automatic modeling approach of three dimensional faces based on phase deviation scanning |
| CN202523112U (en) * | 2012-04-09 | 2012-11-07 | 北京拓博尔机器人科技有限公司 | System for reconstructing three-dimensional model by robot carrying laser scanner |
| CN104036546A (en) * | 2014-06-30 | 2014-09-10 | 清华大学 | Method for carrying out face three-dimensional reconstruction at any viewing angle on basis of self-adaptive deformable model |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI821027B (en) * | 2022-11-17 | 2023-11-01 | 英業達股份有限公司 | Component matching and reporting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106323167A (en) | 2017-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106323167B (en) | A kind of intelligent scanning on-line measurement system and measurement method based on image recognition | |
| CN106091931B (en) | A kind of adaptive scanning measuring system and its control method based on threedimensional model | |
| CN110524581A (en) | A kind of flexible welding robot system and its welding method | |
| CN108109174A (en) | A kind of robot monocular bootstrap technique sorted at random for part at random and system | |
| CN111595949B (en) | Laser ultrasonic imaging detection system and detection method for self-adaptive irregular surface | |
| CN107678307A (en) | Half-practicality imitation test method, apparatus and system | |
| CN110421562A (en) | Mechanical arm calibration system and scaling method based on four item stereo visions | |
| CN106600681A (en) | A method for polishing a curved surface having obstacles | |
| US20190193268A1 (en) | Robotic arm processing system and method, and non-transitory computer-readable storage medium therefor | |
| CN104002602B (en) | The laser activation device and laser activation method of function are corrected with machining accuracy | |
| CN102840825B (en) | Particle locating system and method | |
| CN108480101A (en) | A kind of spray painting control method and apparatus of vision-based detection workpiece identification | |
| CN115290001A (en) | Product testing method and system | |
| CN114289934B (en) | Automatic welding system and method for large structural part based on three-dimensional vision | |
| CN205748263U (en) | The online detection unit of artificial intelligence's sheet metal component production system | |
| CN106903663A (en) | A kind of positioning and marking method, the apparatus and system of the built-in part of revolving shell | |
| CN114295728A (en) | Ultrasonic three-dimensional tomography method for internal defects of complex curved surface workpiece | |
| CN110196231A (en) | A kind of laser-ultrasound off-line detection device and method increasing material product | |
| CN108235769A (en) | The performance test methods and test device of a kind of face recognition device | |
| CN109848989A (en) | A kind of robot execution end automatic Calibration and detection method based on ruby probe | |
| CN108845663A (en) | Design and construction system | |
| CN113850280A (en) | Intelligent airplane part code spraying identification method and device, storage medium and equipment | |
| CN107272664B (en) | Smart tracking control algorithm suitable for photoelectric follow-up verifies system and method | |
| CN105866752A (en) | Radar target track simulator based on microprocessor, and verification testing system | |
| CN117006972A (en) | Object detection method, device, equipment, three-dimensional scanning system and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |